Full-life-cycle data automatic acquisition and statistical analysis method

Abstract

The invention relates to the field of power battery recovery and evaluation, in particular to a full-life-cycle data automatic acquisition and statistical analysis method. Comprising the following steps: acquiring battery identification information and predicted arrival time of a to-be-processed battery; a pre-fetching task is automatically triggered according to the predicted arrival time before arrival, a remote data source is accessed according to a supplier access table by taking battery identification information as a key, and associated historical operation abstract data (including an agreed abstract value set) is acquired and locally cached; identifying a target battery identifier after arrival, and locally retrieving historical operation abstract data of the target battery identifier; collecting real-time detection data on site; and combining historical operation abstract data and real-time detection data analysis, and outputting a residual value level, a security level and / or a shunting instruction. The method has the advantages that automatic acquisition, association and deep analysis of the full life cycle data of the battery are realized, the accuracy and efficiency of residual value evaluation and safety grading of the retired battery are improved, and the problems of data isolation and high misjudgment risk in the prior art are effectively solved.

Description

Technical Field

[0001] This application relates to the field of power battery recycling and evaluation, and more specifically, to a method for automatic data collection and statistical analysis throughout the entire life cycle of batteries. Background Technology

[0002] In modern, large-scale power battery recycling and processing centers, the sorting and pre-processing line handling massive quantities of retired battery packs is the most critical physical checkpoint in the entire recycling value chain. This line processes retired batteries from all over the country daily, representing a diverse range of brands and specifications. However, when these batteries are unloaded from the logistics system and enter the conveyor belt, a significant technological gap exists between the physical movement of the batteries and the synchronization of their digital information. Although the line is equipped with automatic barcode scanners, high-precision vision sensors, and OCV / IMP rapid detection equipment, the current-state data generated by these devices only reflects the physical characteristics of the battery at that specific moment. This method of judgment based solely on current static data is akin to a doctor diagnosing a critically ill patient's entire condition based solely on a thermometer without any medical records—a method fraught with a high risk of misdiagnosis.

[0003] In practice, technicians discovered a serious hidden problem: many batteries that showed normal voltage and internal resistance during incoming inspection had actually undergone multiple severe overcharging, over-discharging, or high-temperature immersion during previous vehicle operation, resulting in irreversible damage such as dried-out electrolyte in the internal cells or closed micropores in the separator. If these batteries were classified as Grade A products with high reuse value based solely on current static data collected on the production line and used in subsequent reuse assembly, it would not only create quality risks for subsequent products but also potentially lead to combustion during disassembly due to internal stress release, posing a serious safety hazard. Conversely, directly crushing and recycling internally healthy batteries simply because of external wear would result in enormous resource waste and economic losses.

[0004] Therefore, the core technical challenge in this scenario lies in the fact that existing data acquisition systems for recycling production lines are isolated and fragmented, lacking a mechanism to automatically correlate and align the battery's current detection data with past operational records within milliseconds. Specifically, the technical challenge is how to automatically read the physical identifiers of the batteries the moment they are brought online, and how to index and extract key characteristics from their previous usage period (such as historical maximum temperature rise, cumulative throughput, and SOH decay trajectory). This data must then be statistically analyzed in conjunction with real-time visual appearance data and electrical performance test data from the production line to quickly calculate the battery's true residual value and safety level. This information can then guide the robotic arm to automatically divert the batteries to the secondary utilization area or the scrapping and crushing area.

[0005] Under the conditions of high-throughput and heterogeneous sources in recycling operations, how to utilize the time window from battery shipment to arrival at the production line to pre-process and extract features from previous operation records, and accurately match and compare the pre-processed results with real-time on-site detection data within milliseconds of the battery scanning process, thereby achieving accurate residual value assessment and safety classification, is a pressing technical problem that needs to be solved. This involves several specific challenges, including: how to determine the information to be pre-fetched during the shipment list stage and allocate limited computing resources to complete protocol parsing and feature extraction; how to provide a deterministic access and parsing process when facing different manufacturers and different administrative access protocols; how to set local cache indexing, validity, and priority rules to ensure millisecond-level retrieval; and how to use rapid on-site detection and pre-prepared degradation logic to complete reliable determination when pre-fetching fails or records are incomplete.

[0006] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0007] The purpose of this application is to provide a method for automatic data collection and statistical analysis throughout the entire battery lifecycle. This method enables automatic collection, correlation, and in-depth analysis of battery lifecycle data, significantly improving the accuracy and efficiency of residual value assessment and safety classification of retired batteries. It also effectively solves the problems of isolated data and high risk of misjudgment in existing technologies.

[0008] This application provides a method for automatic data collection and statistical analysis throughout the entire data lifecycle, the technical solution of which is as follows: The methods include: Obtain battery identification information and estimated arrival time for multiple batteries to be processed, including the target battery to be processed; Based on the estimated arrival time, before the batteries to be processed arrive at the work site, a pre-fetching task is automatically triggered. Using the identification information of each battery as the retrieval key, the remote data source is accessed according to the preset supplier access table to obtain the historical operation summary data generated during the service life of the corresponding battery to be processed. The historical operation summary data is then associated with the corresponding battery identification information and stored in local storage. The historical operation summary data includes the set of agreed summary values ​​returned by the remote data source. In response to the detection that the target battery to be processed has arrived at the work site, the target battery to be processed is identified to obtain its battery identification information; based on the battery identification information of the target battery to be processed, the corresponding historical operation summary data is retrieved from the local storage as the target historical operation summary data; Automatically collects real-time detection data of the target batteries to be processed at the work site; Analyze the target's historical operation summary data and real-time detection data to obtain the statistical analysis results of the target battery to be processed. The statistical analysis results include at least one or more of the following: residual value level, safety level, and shunt command.

[0009] Furthermore, a prefetching task is automatically triggered, using the identification information of each battery as the retrieval key to access remote data sources according to a preset supplier access table, including: For each battery identification information and its corresponding estimated arrival time, the difference between the estimated arrival time and the current system time is calculated to determine the remaining time window; When the remaining time window is greater than the preset time threshold, the battery identification information is added to the first preprocessing queue, and the remote data source is accessed according to the preset first resource configuration parameters. When the remaining time window is less than or equal to the preset time threshold, the battery identification information is added to the second preprocessing queue, and the remote data source is accessed according to the preset second resource configuration parameters. The first resource configuration parameter and the second resource configuration parameter include at least the maximum number of concurrent connections, the request timeout time, and the queue priority identifier, and the priority of the second resource configuration parameter is higher than that of the first resource configuration parameter.

[0010] Furthermore, by analyzing the target's historical operational summary data and real-time detection data, statistical analysis results of the target batteries to be processed are obtained, including: Determine at least one electrical performance indicator to be tested based on real-time detection data; Based on the target's historical operation summary data, determine the historical reference index corresponding to the electrical performance index to be tested. The historical reference index includes the summary characterization value of the electrical performance index to be tested during its service period or its reference range. The electrical performance index to be tested is compared with historical reference indexes to obtain historical comparison results; Obtain the testable electrical performance index samples of the remaining batteries that belong to the same expected arrival batch as the target battery and have been tested from multiple batteries to be processed, and calculate the dynamic statistical benchmark. The dynamic statistical benchmark includes at least the sample mean and sample standard deviation. Based on the degree of deviation between the electrical performance index to be tested and the sample mean, and after normalization according to the sample standard deviation, the standardized deviation value of the target battery to be processed is calculated, and the absolute deviation of the target battery to be processed is calculated based on the electrical performance index to be tested and the sample mean. Based on the type of electrical performance index to be tested and the batch information corresponding to the target battery to be processed, the standardized deviation judgment threshold and the absolute deviation judgment threshold are obtained from the preset judgment threshold table. Based on the type of the electrical performance index to be tested, the historical comparison judgment threshold is obtained from the judgment threshold table. When the absolute value of the standardized deviation is less than the standardized deviation judgment threshold, or the absolute deviation amount is less than the absolute deviation judgment threshold, and the historical comparison results meet the historical comparison judgment threshold, a grade analysis report of the target battery to be processed is generated as the statistical analysis result. When the absolute value of the standardized deviation is greater than or equal to the standardized deviation judgment threshold and the absolute deviation amount is greater than or equal to the absolute deviation judgment threshold, or when the historical comparison result does not meet the historical comparison judgment threshold, an anomaly interception command is generated as a statistical analysis result.

[0011] Furthermore, using the battery identification information as the retrieval key, a remote data source is accessed according to a preset supplier access table to obtain the historical operation summary data generated during the usage period of the corresponding battery to be processed. This historical operation summary data is then associated with the corresponding battery identification information and stored in local storage, including: Query the supplier access table and determine the corresponding remote data source access parameters based on the identification information of each battery. The remote data source access parameters include at least the interface address, authentication method and data format. A data request is sent to the remote data source based on the remote data source access parameters. The data request is used to instruct the remote data source to return a set of agreed-upon summary values. The set of agreed-upon summary values ​​includes at least the quantile description of the voltage rebound characteristic index, the historical maximum temperature rise, and the cumulative throughput. Based on the pre-defined field mapping rules and value normalization rules in the supplier access table, the agreed summary value set is standardized to obtain standardized historical operation summary data. The standardized historical operation summary data is associated with the corresponding battery identification information and stored in local storage. The standardized historical operation summary data is associated with the record collection of timestamp, version number and signature verification value to verify the timeliness and completeness of the standardized historical operation summary data.

[0012] Furthermore, the method also includes: when retrieving target historical operation summary data from local storage fails or the retrieved target historical operation summary data is incomplete, a degradation statistical analysis process is initiated: real-time electrical performance parameters obtained by performing a short-time pulse load test on the target battery to be processed; The deviation is determined based on real-time electrical performance parameters and dynamic statistical benchmarks; Query the preset threshold adjustment table to obtain the threshold judgment threshold corresponding to the expected arrival batch to which the target battery to be processed belongs; When the deviation is greater than or equal to the threshold judgment threshold, an abnormal interception command is generated as the statistical analysis result of the target battery to be processed; When the deviation is less than the threshold, a downgrade level analysis report is generated as the statistical analysis result of the target battery to be processed.

[0013] Furthermore, the method also includes: When the supplier access table does not match the battery identification information, or the remote data source does not return the agreed summary value set within the preset waiting time, the general access and parsing process is executed: Send a general data request to a remote data source to obtain response data. Based on the preset general field location rules, extract the fields from the response data returned by the remote data source to obtain the field values. Based on the preset minimum unit conversion rules, the field values ​​are converted and formatted to generate at least one historical operation summary data item to form historical operation summary data. The historical operation summary data is then associated with the corresponding battery identification information and stored in local storage. When the general access and parsing process fails to obtain historical operation summary data that meets the preset integrity requirements, a field compensation mark is generated for the corresponding battery identification information. After the target battery to be processed arrives at the work site, the detection equipment is controlled to perform a short-time pulse load test on the target battery to supplement the generation of field compensation historical operation summary data items. Based on the field compensation historical operation summary data items, the historical operation summary data associated with the battery identification information and marked with field compensation is compensated and updated to obtain the compensated historical operation summary data. After establishing an association between the compensated historical operation summary data and the battery identification information, it is updated and stored in the local storage.

[0014] Furthermore, the historical operation summary data and real-time detection data of the target are analyzed to obtain the statistical analysis results of the target batteries to be processed, including: obtaining the packing rules of multiple batteries to be processed, which are used to indicate the loading topology of the batteries to be processed in the transport box; Based on the packing rules and the battery identification information of the target battery to be processed, the target battery to be processed is mapped to the hot zone inside the transport box. The hot zone includes at least the edge zone, the transition zone and the core zone. The expected temperature deviation level of the target battery to be treated is determined based on the thermal location using a preset temperature response coefficient table. Based on the temperature response coefficient table and the expected temperature deviation level, the real-time detection data is normalized and corrected to obtain the temperature-corrected real-time detection data. The statistical analysis results of the target battery to be processed are obtained by analyzing the real-time detection data after temperature correction and the target historical operation summary data.

[0015] Furthermore, the real-time detection data includes visual appearance data. Analysis of the target's historical operational summary data and real-time detection data yields statistical analysis results for the target battery to be processed, including: When visual appearance data indicates that the target battery to be processed has physical defects, determine the type and location of the physical defects; Perform a targeted search in the target historical operation summary data to find stress event summaries that occurred within a preset time window and whose amplitude exceeded a preset physical threshold. The stress event summary includes at least the stress event occurrence time and stress event amplitude. When a stress event summary that matches a physical defect with a preset matching condition is retrieved, it is determined that the target battery to be treated has an internal damage risk, and a statistical analysis result containing a high-risk safety level and an abnormal interception command is generated.

[0016] Furthermore, the method also includes: Obtain the transportation trajectory timeline of multiple batteries to be processed, which includes the location and time information of the multiple batteries to be processed; Before the target battery arrives at the work site, the transportation environment summary data of the target battery is generated by accessing third-party meteorological data sources or vehicle sensor records based on the transportation trajectory timeline. After analyzing the target's historical operation summary data and real-time detection data to obtain the statistical analysis results of the target battery to be processed, the process also includes: determining the predictive reference range for consistency judgment based on the transportation environment summary data and the target's historical operation summary data, and performing consistency judgment between the real-time detection data and the predictive reference range to obtain the consistency judgment result; When the consistency determination result is within the preset consistency threshold range, the target battery to be processed is determined to be in a state that meets expectations during transportation, and the statistical analysis results are corrected based on the consistency determination result. When the consistency determination result exceeds the consistency threshold range, the statistical analysis result will be marked as transportation loss pending.

[0017] Furthermore, a summary of the transportation environment data for the corresponding target battery to be processed is generated, including: The transportation process is divided into multiple consecutive time periods based on the transportation trajectory timeline; For each time period, based on the location and time information corresponding to the transportation trajectory timeline, environmental data corresponding to that time period is obtained from third-party meteorological data sources or vehicle-mounted sensor records. The environmental data is summarized to generate transportation environmental summary data. The transportation environmental summary data includes at least one or more of the following: the highest temperature, the lowest temperature, the temperature change range, and the cumulative duration of the temperature exceeding the preset temperature threshold during the transportation process. The transportation environment summary data is associated with the corresponding battery identification information and stored in local storage for consistency determination.

[0018] As can be seen from the above, the method for automatic data collection and statistical analysis throughout the entire battery lifecycle provided in this application automatically triggers a pre-fetching task before the battery arrives at the work site by acquiring the identification information and estimated arrival time of the battery to be processed, accessing a remote data source to obtain historical operation summary data and storing it locally; when the target battery to be processed arrives at the work site, its identification information is identified, historical data is retrieved from local storage, and real-time detection data is automatically collected; finally, the historical data and real-time data are comprehensively analyzed to obtain statistical analysis results including residual value level, safety level, and shunt command. This method has the advantages of realizing automatic collection, correlation, and in-depth analysis of battery lifecycle data, significantly improving the accuracy and efficiency of residual value assessment and safety classification of retired batteries, and effectively solving the problems of data isolation and high risk of misjudgment in the prior art. Attached Figure Description

[0019] Figure 1 This application provides a flowchart illustrating a method for automatic data collection and statistical analysis throughout the entire data lifecycle. Detailed Implementation

[0020] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0021] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0022] Reference Figure 1 This application proposes a method for automatic data collection and statistical analysis throughout the entire data lifecycle, including: S1000: Obtain battery identification information and estimated arrival time of multiple batteries to be processed, including the target battery to be processed; S2000: Based on the estimated arrival time, before the battery to be processed arrives at the work site, the pre-fetching task is automatically triggered. Using the identification information of each battery as the retrieval key, the remote data source is accessed according to the preset supplier access table to obtain the historical operation summary data formed during the service life of the corresponding battery to be processed. The historical operation summary data is then associated with the corresponding battery identification information and stored in the local storage. The historical operation summary data includes the set of agreed summary values ​​returned by the remote data source. Specifically, the supplier access table should include at least the interface address, authentication method, data format, and field mapping / unit conversion rules; S3000: In response to the detection that the target battery to be processed has arrived at the work site, the target battery to be processed is identified to obtain its battery identification information; S4000: Using the battery identification information of the target battery as the search key, retrieve the corresponding historical operation summary data from the local storage as the target historical operation summary data; S5000: Automatically collects real-time detection data of the target battery to be processed at the work site; S6000: Analyze the target's historical operation summary data and real-time detection data to obtain the statistical analysis results of the target battery to be processed. The statistical analysis results include at least one or more of the following: residual value level, safety level, and shunting command.

[0023] In this embodiment, firstly, the battery identification information and estimated arrival time of multiple batteries to be processed are obtained, including the target battery to be processed. This step is the starting point of the entire data preprocessing process. When the logistics system generates a shipping list containing information about the batteries to be recycled, such as an electronic data stream containing fields such as unique battery identifiers, estimated shipping time, estimated arrival time, and shipper name, the data processing device immediately receives and parses this list. It then extracts the battery identification information and estimated arrival time corresponding to each battery to be processed from the electronic data stream, forming the input set and time reference for subsequent pre-fetching tasks.

[0024] Next, based on the estimated arrival time, a pre-fetching task is automatically triggered before the batteries arrive at the work site. This pre-fetching task uses each battery's identification information as the search key to access the remote data source according to a pre-defined supplier access table to obtain the historical operational summary data generated during the battery's usage period. The obtained historical operational summary data is then associated with the corresponding battery identification information and stored in local storage. Specifically, the historical operational summary data represents a set of agreed-upon summary values ​​returned by the remote data source. The supplier access table provides the interface address, authentication method, data format, and field mapping / unit conversion rules for remote access, allowing batteries from different sources to determine their access path and parsing method according to the table. Local storage caches the historical operational summary data corresponding to the battery identification information, facilitating quick retrieval using the search key later. By completing remote access and summary data preparation before the batteries arrive, the risk of waiting due to accessing the remote data source after arrival is reduced.

[0025] When information indicating that a target battery has arrived at the work site is detected—for example, when a battery pack is placed on a conveyor belt and passes through an automatic barcode scanner or high-frequency wireless identification device at the production line entrance—the target battery is identified to obtain its battery identification information. This battery identification information is used to establish a consistent retrieval key with historical operation summary data cached locally, thereby achieving a one-to-one match between the target battery and its historical operation summary data.

[0026] Subsequently, using the battery identification information of the target battery as the search key, the corresponding historical operation summary data is retrieved from local storage as the target historical operation summary data. Since the data has been pre-acquired and stored on local high-speed storage media, such as solid-state drive arrays or in-memory databases, rapid retrieval can be completed under production line cycle time requirements. The target historical operation summary data is used as historical input for subsequent statistical analysis.

[0027] Meanwhile, real-time detection data of the target batteries to be processed is automatically collected at the work site. This data is generated by various rapid detection devices deployed on the production line, such as high-precision vision sensors to capture appearance defects, and rapid detection devices for open-circuit voltage and AC internal resistance to obtain instant electrical performance parameters. The real-time detection data is used as on-site input for subsequent statistical analysis.

[0028] Finally, the retrieved historical operational summary data and real-time monitoring data collected on-site are analyzed to obtain statistical analysis results for the target batteries to be processed. These results include at least one or more of the following: residual value level, safety level, and diversion command. This is a comprehensive decision-making process that comprehensively assesses the health status of the battery by comparing its past and present conditions. For example, the residual value level can be divided into A, B, and C levels, corresponding to secondary use, material recycling, or direct scrapping, respectively; the safety level can be divided into low, medium, and high risk; and the diversion command is a direct command that drives the robotic arm or sorting device on the production line to perform physical operations, such as sorting the battery to the secondary use buffer area or the scrap crushing conveyor belt. In this way, this method connects the entire lifecycle data stream of the battery, achieving closed-loop automation from data acquisition to physical sorting, significantly improving recycling efficiency, accuracy, and safety.

[0029] In another embodiment of this application, an automatic prefetching task is proposed, using each battery identification information as the retrieval key to access a remote data source according to a preset supplier access table, including: S2100: For each battery identification information and its corresponding estimated arrival time, calculate the difference between the estimated arrival time and the current system time to determine the remaining time window; S2200: When the remaining time window is greater than the preset time threshold, add the battery identification information to the first preprocessing queue and access the remote data source according to the preset first resource configuration parameters; S2300: When the remaining time window is less than or equal to the preset time threshold, add the battery identification information to the second preprocessing queue and access the remote data source according to the preset second resource configuration parameters; S2400: Wherein, the first resource configuration parameter and the second resource configuration parameter include at least the maximum number of concurrent connections, the request timeout time and the queue priority identifier, and the priority of the second resource configuration parameter is higher than the priority of the first resource configuration parameter.

[0030] To manage and execute prefetch tasks more effectively, especially when faced with a large number of concurrent tasks and limited computing resources, a dynamic scheduling strategy based on time urgency can be introduced. For example, the preprocessing service maintains two task queues (a first preprocessing queue and a second preprocessing queue), and the scheduler fetches tasks for execution from high to low queue priority.

[0031] This strategy calculates the difference between the estimated arrival time and the current system time for each battery identification information and its corresponding estimated arrival time, thereby determining the remaining time window. The remaining time window measures the available time to complete the prefetch and is used as the basis for prioritizing tasks. For example, if the current system time is 10:00 and the estimated arrival time is 02:00 the next day, then the remaining time window is 16 hours; if the estimated arrival time is 18:00 on the same day, then the remaining time window is 8 hours.

[0032] When the remaining time window exceeds a preset time threshold (e.g., more than twelve hours), the battery identification information is added to the first preprocessing queue, and the remote data source is accessed according to the first resource configuration parameters. The first resource configuration parameters can adopt a relatively conservative resource strategy to smooth the load and reduce remote pressure, such as reducing the upper limit of concurrent connections, appropriately extending the request timeout time, and setting a lower queue priority flag. For example, the first resource configuration parameters can be configured as follows: upper limit of concurrent connections = 10, request timeout time = 5 seconds, queue priority flag = P1; the scheduler takes N tasks from the first preprocessing queue every second for batch execution, and initiates a request according to the interface address and authentication method selected from the supplier access table to obtain the agreed digest value set.

[0033] When the remaining time window is less than or equal to a preset time threshold (e.g., less than or equal to twelve hours), the battery identification information is added to the second preprocessing queue, and the remote data source is accessed according to the second resource configuration parameters. The second resource configuration parameters correspond to a higher execution priority, with a queue priority identifier higher than the queue priority identifier of the first resource configuration parameters. It can also increase the maximum number of concurrent connections and shorten the request timeout time to prioritize urgent tasks. For example, the second resource configuration parameters can be configured as follows: maximum number of concurrent connections = 50, request timeout time = 2 seconds, queue priority identifier = P0. The scheduler prioritizes retrieving tasks from the second preprocessing queue, and if necessary, postpones retrieval from the first preprocessing queue to ensure that batteries arriving closer to the deadline are pre-fetched first.

[0034] By employing a differentiated scheduling mechanism with dual queues and dual parameters, the timeliness of prefetching can be improved, and the risk of on-site waiting caused by incomplete prefetching before arrival can be reduced. The preset time threshold, the first resource configuration parameter, and the second resource configuration parameter can be maintained by a configuration table (e.g., a database table or configuration file) and can be adjusted according to production line throughput and remote interface rate limiting policies. For example, during peak periods, the maximum concurrent connection count of the first preprocessing queue can be reduced to 5 to decrease remote pressure, while keeping the maximum concurrent connection count of the second preprocessing queue unchanged.

[0035] In another embodiment of this application, it is further proposed that, using the battery identification information as the retrieval key, a remote data source is accessed according to a preset supplier access table to obtain the historical operation summary data generated during the usage period of the corresponding battery to be processed, and the historical operation summary data is associated with the corresponding battery identification information and stored in local storage, including: S2500: Query the supplier access table and determine the corresponding remote data source access parameters based on the identification information of each battery. The remote data source access parameters include at least the interface address, authentication method and data format. S2600: Sends a data request to the remote data source based on the remote data source access parameters. The data request is used to instruct the remote data source to return a set of agreed-upon summary values. The set of agreed-upon summary values ​​includes at least the quantile description of the voltage rebound characteristic index, the historical maximum temperature rise, and the cumulative throughput. S2700: Based on the preset field mapping rules and value normalization rules in the supplier access table, the agreed summary value set is standardized to obtain standardized historical operation summary data; S2800: After establishing an association between the standardized historical operation summary data and the corresponding battery identification information, it stores the data in local storage. It also records the timestamp, version number, and signature verification value associated with the standardized historical operation summary data to verify the timeliness and completeness of the standardized historical operation summary data.

[0036] In the data prefetching stage, in order to solve the problem of different data interfaces, protocols and formats of different battery suppliers or operators, and to ensure that the acquired data is standardized and usable, a set of deterministic access, parsing and standardization processes are included. The process uses the identification information of each battery as the retrieval key, accesses the remote data source according to the preset supplier access table, obtains the historical operation summary data generated during the service life of the corresponding battery to be processed, and stores the historical operation summary data in local storage after establishing an association with the corresponding battery identification information.

[0037] First, the supplier access table is queried to determine the remote data source access parameters based on each battery's identification information. Access parameters include at least the interface address, authentication method, and data format. Supplier matching can be achieved through fields such as the battery identification information itself, the shipper's name, or the operator's code in the shipping manifest. In the example, when the battery comes from supplier A, the interface address can be https: / / api.vendorA.com / v1 / battery, the authentication method is OAuth2.0, and the data format is JSON; when the battery comes from supplier B, the interface address can be http: / / data.vendorB.cn / query, the authentication method is an API key (carrying a key field in the request header), and the data format is XML; when the battery comes from supplier C, the interface address can be a SOAP endpoint, the authentication method is username and password, and the data format is SOAP-encapsulated XML.

[0038] After determining the access parameters, a data request is sent to the remote data source. The data request instructs the remote data source to return a predefined summary set of values, rather than the complete raw time-series data, to reduce transmission and parsing overhead. The predefined summary set of values ​​includes at least a quantile description of the voltage rebound characteristic index, the historical maximum temperature rise, and the cumulative throughput. An example request can be organized using a parameterized approach: battery_id=BAT001&features=voltage_rebound_quantile, `max_temp_rise,throughput`, where `features` is used to limit the set of returned fields. Quantile descriptions can be combinations of values ​​for discrete quantiles such as 25%, 50%, and 75%. The meaning and value scope of the historical maximum temperature rise field are defined by the supplier access table; for example, it can be defined as the highest temperature or the highest temperature rise during operation, facilitating semantic alignment of data from different sources.

[0039] After the remotely returned set of agreed-upon summary values, standardization processing is required. Standardization can include both field mapping rules and value normalization rules, both of which are pre-defined in the supplier access table. Field mapping unifies different supplier field names to internal field names; for example, `temp_max_celsius` and `BatteryMaxTemp` are mapped to the historical maximum temperature rise. Value normalization unifies units and dimensions, such as converting Fahrenheit to Celsius, milliampere-hours to ampere-hours, and millivolts to volts. For instance, if supplier B returns data in XML... <maxtempf> 131< / maxtempf> The temperature can be converted to 55℃ according to the rules and written into the historical maximum temperature rise field; if the returned throughput is 120000mAh, it can be converted to 120Ah and written into the cumulative throughput field. After completion, standardized historical operation summary data is obtained, which is convenient for subsequent cross-supplier unified comparison.

[0040] Finally, the standardized historical operation summary data is associated with the battery identification information and stored locally. To ensure timeliness and integrity, the collection timestamp, version number, and signature verification value can be recorded synchronously. The collection timestamp is used to determine data freshness and cache validity; the version number is used to distinguish multiple prefetches or updates of the same battery (e.g., incrementing by the number of prefetches, or using the remotely returned version field); the signature verification value is used for tamper-proof verification, such as calculating the SHA-256 hash verification value of the summary data content, or recording the digital signature verification result returned from the remote end, thereby ensuring that the locally stored data is traceable and verifiable. During the prefetching process, situations may arise where the supplier access table cannot match the battery identification information, or the remote data source fails to return the agreed summary value set within the preset waiting time. To address these anomalies, a degradation processing and on-site compensation mechanism has been designed.

[0041] In another embodiment of this application, the method further includes: S2900: When the supplier access table does not match the battery identification information, or the remote data source does not return the agreed summary value set within the preset waiting time, the general access and parsing process is executed: a general data request is sent to the remote data source to obtain response data; according to the preset general field position rules, the fields returned by the remote data source are extracted to obtain field values; according to the preset minimum unit conversion rules, the field values ​​are converted and formatted to generate at least one historical operation summary data item to constitute historical operation summary data; and the historical operation summary data is associated with the corresponding battery identification information and stored in local storage. S21000: When the general access and parsing process does not obtain historical operation summary data that meets the preset integrity requirements, a field compensation mark is generated for the corresponding battery identification information. After the target battery to be processed arrives at the work site, the detection equipment is controlled to perform a short-time pulse load test on the target battery to be processed in order to supplement the generation of field compensation historical operation summary data items. Based on the field compensation historical operation summary data items, the historical operation summary data associated with the battery identification information and with the field compensation mark is compensated and updated to obtain the compensated historical operation summary data. After establishing an association between the compensated historical operation summary data and the battery identification information, it is updated and stored in the local storage.

[0042] When the supplier access table does not match the battery identification information, or the remote data source fails to return the agreed-upon summary value set within the preset waiting time, the general access and parsing process will be executed first. The preset waiting time can be used as a timeout threshold for remote requests, such as 300 milliseconds, 500 milliseconds, or 1 second. If no valid response is received after the timeout threshold is reached, the dedicated access will end and the general process will begin, avoiding blocking the prefetch queue.

[0043] The general access and parsing process attempts a data acquisition method that does not rely on specific vendor protocols. General data requests can use pre-defined general request templates, such as " / battery / basic?sn=xxx" or " / device / telemetry / latest?id=xxx" from a REST interface, or they can use a general query entry provided by the platform. The request templates are fixed in the system configuration, ensuring process determinism. After obtaining the response data, fields are extracted according to pre-defined general field location rules to obtain field values. General field location rules can be defined using a "candidate path list + priority" approach. For example, for JSON, paths such as data.battery.voltage, data.pack.voltage, and battery.voltageV are tried sequentially; for XML, XPath paths such as / / battery / voltage and / / pack / voltage are tried sequentially. When a path is matched and the value meets the format requirements, the matched path is recorded and the field value is output.

[0044] After extracting the field values, unit conversion and format standardization are performed according to preset minimum unit conversion rules, such as converting millivolts to volts. The minimum unit conversion rules cover common units: mV → V (value / 1000), mΩ → Ω (value / 1000), mAh → Ah (value / 1000), Fahrenheit → Celsius ((F-32) / 1.8). Simultaneously, the numerical format is standardized to fixed decimal places or scientific notation to avoid ambiguity in subsequent parsing. In this way, even without dedicated parsing rules, at least one historical operation summary data item can be generated to constitute historical operation summary data. The historical operation summary data generated here must at least include this historical operation summary data item; other summary items can be omitted and stored locally after being associated with the battery identification information. "Source = General Access and Parsing Process" and "Hit Path / Conversion Rule" can also be recorded as audit information.

[0045] The general access and parsing process cannot guarantee that historical operational summary data meeting the preset integrity requirements will be obtained. If the preset integrity requirements are still not met after the process is completed, such as missing key summary items, a field compensation flag will be generated for the corresponding battery identification information. The preset integrity requirements can adopt deterministic criteria, such as: the historical operational summary data must contain at least two of the following: "quantile description of voltage rebound characteristic index, historical maximum temperature rise, and cumulative throughput", or at least the key summary item "quantile description of voltage rebound characteristic index"; if not met, it is judged as incomplete and a field compensation flag is written. The field compensation flag can be saved as a status field of a local storage entry, such as status=NEED_ONSITE_COMPENSATION, to indicate that the historical data file of the battery has defects and additional supplementary testing is required after the physical entity arrives at the work site.

[0046] When a target battery marked with a field compensation tag arrives at the work site, a compensation process is triggered. The testing equipment performs a short-duration pulse load test on the target battery to supplement the generated field compensation historical operation summary data. For example, a discharge pulse of 10 amps for 100 milliseconds is applied to the battery. After the pulse ends, the voltage recovery curve is collected and the voltage rebound slope is calculated as a field compensation historical operation summary data item. Alternatively, open-circuit voltage and internal resistance measurements can be collected simultaneously as supplementary items. Subsequently, based on the field compensation historical operation summary data items, the historical operation summary data associated with the battery identification information and marked with a field compensation tag in the local storage is updated to obtain the compensated historical operation summary data. The compensation update can adopt a field-level merging strategy: priority is given to filling missing summary items; if a field already exists but is marked as invalid or abnormal, it can be updated incrementally by version number, and the differences before and after the update are recorded. Finally, the compensated historical operation summary data is associated with the battery identification information and updated and stored in local storage for use in subsequent analysis processes. By employing a two-stage degradation and compensation mechanism, the robustness of data acquisition is improved, ensuring that a usable data profile can still be established for each battery even under adverse conditions.

[0047] In another embodiment of this application, it is further proposed to analyze the target's historical operation summary data and real-time detection data to obtain statistical analysis results of the target battery to be processed, including: S6100: Determine at least one electrical performance indicator to be tested based on real-time detection data; S6200: Based on the target's historical operation summary data, determine the historical reference index corresponding to the electrical performance index to be tested. The historical reference index includes the summary characterization value of the electrical performance index to be tested during its service life or its reference range. S6300: Compares the electrical performance index to be tested with historical reference indexes to obtain historical comparison results; S6400: Obtain the test electrical performance index samples of the remaining batteries to be processed from multiple batteries to be processed that belong to the same expected batch of batteries to be processed and have been tested, and calculate the dynamic statistical benchmark. The dynamic statistical benchmark includes at least the sample mean and sample standard deviation. S6500: Based on the degree of deviation between the electrical performance index under test and the sample mean, and after normalization according to the sample standard deviation, calculate the standardized deviation value of the target battery to be processed, and calculate the absolute deviation of the target battery to be processed based on the electrical performance index under test and the sample mean. S6600: Based on the type of electrical performance index to be tested and the batch information corresponding to the target battery to be processed, the standardized deviation judgment threshold and the absolute deviation judgment threshold are obtained from the preset judgment threshold table. S6700: Based on the type of the electrical performance index to be tested, the historical comparison judgment threshold is obtained from the judgment threshold table. S6800: When the absolute value of the standardized deviation is less than the standardized deviation judgment threshold, or the absolute deviation amount is less than the absolute deviation judgment threshold, and the historical comparison results meet the historical comparison judgment threshold, a grade analysis report of the target battery to be processed is generated as a statistical analysis result. S6900: When the absolute value of the standardized deviation is greater than or equal to the standardized deviation judgment threshold and the absolute deviation amount is greater than or equal to the absolute deviation judgment threshold, or when the historical comparison result does not meet the historical comparison judgment threshold, an anomaly interception command is generated as a statistical analysis result.

[0048] Once the target battery arrives at the work site and completes identification, the system can retrieve historical operational summary data locally. Simultaneously, on-site detection equipment collects real-time detection data, and the system enters the statistical analysis phase. The statistical analysis employs a combination of historical longitudinal comparison and batch-specific horizontal statistics, outputting a graded analysis report or anomaly interception instructions.

[0049] First, based on the real-time detection data collected on-site, determine at least one electrical performance indicator to be tested. This indicator may include one or more of the following: voltage rebound characteristic and internal resistance. Example: Obtain the internal resistance value using an AC internal resistance tester; or apply a short-time pulse load to the battery, record the voltage V1 at the 1st second and the voltage V5 at the 5th second after the pulse ends, and calculate the voltage rebound slope as the voltage rebound characteristic indicator using (V1−V5) / 4.

[0050] Next, based on the target historical operating summary data retrieved from local storage, determine the historical reference index corresponding to the electrical performance index under test. The historical reference index can be a summary characterization value (such as median internal resistance, average internal resistance) or a reference range (such as the 25th to 75th quantile range of internal resistance, or the quantile description range of the voltage rebound characteristic index). Example: If the target historical operating summary data records the quantile description of the voltage rebound characteristic index as P25=0.015, P50=0.012, and P75=0.009, then [P25, P75] can be used as the reference range, or P50 can be used as the summary characterization value.

[0051] Then, the electrical performance index to be tested is compared with historical reference indexes to obtain historical comparison results. The historical comparison results can be defined as "satisfied / unsatisfied" according to a threshold table, such as whether the current internal resistance falls within the historical reference range, or whether the deviation between the current voltage rebound characteristic index and the historical summary characterization value exceeds the historical comparison judgment threshold.

[0052] In addition to longitudinal comparisons with its own historical data, the system also needs to perform horizontal comparisons with other batteries in the same expected arrival batch. The system acquires samples of the electrical performance indicators of the remaining batteries in the same expected arrival batch that have already undergone testing, and calculates a dynamic statistical baseline. This dynamic statistical baseline includes at least the sample mean and sample standard deviation. Expected arrival batches can be categorized by expected arrival time, for example, by the same train or the same arrival time window (e.g., 30 minutes or 1 hour). As production line testing progresses, the sample mean and sample standard deviation can be updated incrementally to avoid redundant full-batch calculations.

[0053] Based on a dynamic statistical benchmark, the system calculates two types of deviations. The first type is the standardized deviation, which is the deviation of the measured electrical performance index from the sample mean, normalized to the sample standard deviation, and can be calculated using the Z-score. The second type is the absolute deviation, which is the absolute value of the difference between the measured electrical performance index and the sample mean. Example: If the sample mean of the expected arriving batch of internal resistance is 12mΩ and the sample standard deviation is 2mΩ, and the measured internal resistance of the target battery to be processed is 18mΩ, then the standardized deviation = (18−12) / 2 = 3, and the absolute deviation = |18−12| = 6mΩ.

[0054] To achieve configurable and auditable judgments, the system obtains judgment thresholds from a preset judgment threshold table. Based on the type of the electrical performance indicator to be tested and the batch information corresponding to the target battery, it obtains standardized deviation judgment thresholds and absolute deviation judgment thresholds; it also obtains historical comparison judgment thresholds based on the indicator type. The judgment threshold table can be organized as "indicator type × batch information." For example, the internal resistance indicator can be configured with one set of thresholds (standardized deviation judgment thresholds and absolute deviation judgment thresholds) for a certain expected batch, while the voltage rebound characteristic indicator can be configured with another set of thresholds. Historical comparison judgment thresholds can be given by indicator type; for example, it can be specified that the deviation of the current value relative to the historical P50 must not exceed a preset proportion or a fixed amount.

[0055] The final decision adopts a joint rule of two types of deviation criteria and historical comparison results: when the standardized deviation value is less than the standardized deviation judgment threshold, or the absolute deviation is less than the absolute deviation judgment threshold, and the historical comparison results meet the historical comparison judgment threshold, a grade analysis report of the target battery to be processed is generated as the statistical analysis result; when the standardized deviation value is greater than or equal to the standardized deviation judgment threshold and the absolute deviation is greater than or equal to the absolute deviation judgment threshold, or the historical comparison results do not meet the historical comparison judgment threshold, an anomaly interception instruction is generated as the statistical analysis result, and the production line is instructed to transfer the battery to the anomaly review channel.

[0056] In another embodiment of this application, the method further includes: S61000: When retrieving target historical runtime summary data from local storage fails, or the retrieved target historical runtime summary data is incomplete, the degradation statistical analysis process is initiated: S61100: Acquire real-time electrical performance parameters obtained by performing a short-time pulse load test on the target battery to be processed; S61200: Determine the deviation based on real-time electrical performance parameters and dynamic statistical benchmarks; S61300: Query the preset threshold adjustment table to obtain the threshold judgment threshold corresponding to the expected arrival batch to which the target battery to be processed belongs; S61400: When the deviation is greater than or equal to the threshold judgment threshold, an abnormal interception command is generated as the statistical analysis result of the target battery to be processed; S61500: When the deviation is less than the threshold judgment threshold, a downgrade level analysis report is generated as the statistical analysis result of the target battery to be processed.

[0057] In some scenarios, even if the prefetch task has been triggered, local storage may still experience retrieval failures or incomplete data, such as incomplete writes due to network transmission interruptions, corrupted cache entries, or abnormal storage media. To avoid interruptions in the main analysis process due to a lack of historical runtime summary data, the system directly switches to a preset degraded statistical analysis process when retrieval fails or integrity checks fail.

[0058] The core of the downgraded statistical analysis process is that it no longer relies on historical operation summary data as an individual baseline, but instead uses real-time electrical performance parameters obtained from short-time pulse load testing as the main basis, and combines them with dynamic statistical benchmarks formed by the samples that have been tested within the expected batch to make a judgment.

[0059] After the process is started, the testing equipment is first controlled to perform a short-time pulse load test on the target battery to obtain real-time electrical performance parameters. The short-time pulse load test can adopt a non-destructive, rapid operating condition, such as applying a discharge pulse of 10 amperes for 100 milliseconds to the battery, and collecting the voltage recovery curve and current curve after the pulse ends, thereby calculating real-time electrical performance parameters such as the measured internal resistance and rebound slope, which serve as the main input in the degradation mode.

[0060] After acquiring real-time electrical performance parameters, the system aggregates the remaining unprocessed battery samples within the expected batch arrival period, calculates dynamic statistical benchmarks (sample mean and sample standard deviation), and calculates the deviation accordingly. The definition of the deviation is consistent with the threshold adjustment table, and can be either absolute deviation (e.g., |X−μ|) or standardized deviation (e.g., (X−μ) / σ). Example: If the batch internal resistance sample mean is 12mΩ, the sample standard deviation is 2mΩ, and the measured internal resistance of the target battery is 18mΩ, then the absolute deviation is 6mΩ, and the standardized deviation is 3.

[0061] When the number of samples that have been tested within the expected batch is insufficient to form a stable statistical baseline, an engineering-feasible compensation method can be adopted, such as extending to samples in adjacent arrival time windows, referencing the statistical baseline of the previous expected batch as the initial value, or marking "insufficient sample size" in the downgrade analysis report and increasing the review priority, so as to ensure that an executable judgment output can still be given under the production line cycle time.

[0062] The preset threshold adjustment table is then consulted to obtain the threshold judgment threshold corresponding to the expected arrival batch. The threshold adjustment table can be configured with thresholds according to "indicator type + deviation type + expected arrival batch". For example, a threshold of 6mΩ can be configured for the absolute deviation of the internal resistance indicator, and a threshold of 3 can be configured for the standardized deviation value of the internal resistance indicator. Different thresholds can also be configured for batches from different sources or with different aging levels, so as to maintain the consistency and auditability of the judgment in the degradation mode.

[0063] Finally, a judgment is made: when the deviation is greater than or equal to the threshold judgment result, an anomaly interception command is generated as a statistical analysis result, used to transfer the battery to the anomaly review channel to prevent high-risk batteries from entering subsequent sorting or secondary utilization stages; when the deviation is less than the threshold judgment result, a downgrade level analysis report is generated as a statistical analysis result. The downgrade level analysis report can clearly indicate whether historical operation summary data is missing or incomplete, the type of deviation, the source of dynamic statistical benchmarks and threshold judgment results, which facilitates subsequent traceability and provides an executable diversion basis for sorting equipment.

[0064] In actual battery recycling operations, the physical environment during transportation, especially temperature, can have a significant temporary impact on the electrochemical performance of batteries. For example, in container transport, batteries near the container walls and those in the center experience different temperature profiles, resulting in different temperatures upon arrival at the site. This affects the accuracy of real-time monitoring data such as internal resistance. To eliminate this measurement bias caused by the transportation environment, a temperature normalization correction process based on the loading topology can be introduced during the analysis of historical operational summary data and real-time monitoring data.

[0065] In another embodiment of this application, it is further proposed to analyze the target's historical operation summary data and real-time detection data to obtain statistical analysis results of the target battery to be processed, including: S6001: Obtain the packing rules for multiple batteries to be processed. The packing rules are used to indicate the loading topology of the batteries to be processed in the transport container. S6002: Based on the packing rules and the battery identification information of the target battery to be processed, map the target battery to be processed to the hot zone inside the transport box. The hot zone includes at least the edge zone, the transition zone and the core zone. S6003: Determine the expected temperature deviation level of the target battery to be processed based on the preset temperature response coefficient table according to the thermal location; S6004: Based on the temperature response coefficient table and the expected temperature offset level, perform temperature normalization correction on the real-time detection data to obtain temperature-corrected real-time detection data. S6005: Analyze the real-time detection data after temperature correction and the target's historical operation summary data to obtain the statistical analysis results of the target battery to be processed.

[0066] The process first obtains the packing rules for multiple batteries to be processed. Packing rules can be structured data, describing information such as the dimensions of the transport container, the number of stacking layers, the row and column layout of each layer, and the slot numbers. The sources of packing rules can be implemented in several ways: for example, at the packing station, a barcode scanner sequentially scans the container number and battery identification information, and the system automatically records the binding relationship according to "layer-row-column / slot number"; or the packing list is imported from the warehousing / logistics system, which directly contains the correspondence between battery identification information and the slots inside the container; or operators can select a container template and enter the loading layout in the configuration interface.

[0067] Subsequently, based on the packing rules and the battery identification information of the target batteries to be processed, the target batteries are mapped to the hot zones within the transport container. The mapping can directly utilize the "battery identification information - slot number" correspondence generated during the packing stage, and then categorize them into edge zones, transition zones, or core zones according to the slot number; the sequence number arrangement can be an optional implementation, but not the sole dependency. Hot zones can be predefined according to the container's geometric location: edge zones correspond to the set of slots closest to the container wall or door, core zones correspond to the set of slots in the center of the container, and transition zones correspond to the set of slots between the two. The mapping process can be completed by rule calculation and does not necessarily rely on additional sensors; if a surface temperature gun or BMS temperature reporting is available on-site, the measured temperature can be incorporated into subsequent corrections to improve accuracy.

[0068] After determining the hot zone location, the expected temperature offset level is obtained from a pre-defined temperature response coefficient table. The temperature response coefficient table can be established based on experimental data or manufacturer specifications, and should at least differentiate between dimensions such as battery type, hot zone location, season, or route type, and provide the binding relationship between the offset level and the temperature offset value. Example: Set offset levels L2=+5℃, L1=+2℃, L0=0℃, L-1=-2℃; in summer long-distance transportation scenarios, L2 can be used for the edge area, and L0 can be used for the core area.

[0069] Next, the real-time detection data collected on-site is normalized and corrected by combining the expected temperature deviation level with the correction coefficients in the temperature response coefficient table. The correction method can be linear conversion: for example, using 25℃ as the standard temperature, when the coefficient table states "internal resistance decreases by 0.05mΩ for every 1℃ increase," if the deviation level corresponds to +5℃, then the equivalent standard temperature internal resistance = measured internal resistance + 0.05 × 5mΩ. Temperature correction coefficients can also be set for indicators such as open-circuit voltage or rebound slope, and normalization can be performed separately to obtain the temperature-corrected real-time detection data.

[0070] Finally, the temperature-corrected real-time monitoring data is jointly analyzed with the target's historical operational summary data to output statistical analysis results. By first eliminating temporary temperature deviations caused by transportation and then comparing them with historical operational summary data, data comparability can be improved, and the risk of misjudgment caused by differences in the temperature of the arriving goods can be reduced. This is especially suitable for scenarios where indicators such as internal resistance and rebound are sensitive to temperature.

[0071] Besides electrical performance data, the physical appearance of a battery is also an important basis for assessing its safety status. Real-time monitoring data typically includes visual appearance data captured by high-resolution industrial cameras. The process of analyzing the target's historical operational summary data and real-time monitoring data also includes a specific analytical logic that cross-references visual defects with historical stress events.

[0072] In another embodiment of this application, it is further proposed that the real-time detection data includes visual appearance data, and that the historical operation summary data of the target and the real-time detection data are analyzed to obtain the statistical analysis results of the target battery to be processed, including: S6010: When visual appearance data indicates that the target battery to be processed has a physical defect, determine the type and location of the physical defect; S6020: Perform a targeted search in the target historical operation summary data to find stress event summaries that occurred within a preset time window and whose amplitude exceeds a preset physical threshold. The stress event summary includes at least the stress event occurrence time and stress event amplitude. S6030: When a stress event summary that matches the physical defect with a preset condition is retrieved, it is determined that the target battery to be treated has an internal damage risk, and a statistical analysis result containing a high-risk safety level and an abnormal interception command is generated.

[0073] When visual appearance data indicates that the target battery has a physical defect, such as an image analysis algorithm detecting a dent, scratch, or bulge on the battery casing, the defect type and location are output first. The defect location can be represented by a camera coordinate system or a battery casing partition number; for example, when taking pictures with a top-view + side-view camera at a conveyor belt station, the pixel positions are converted into casing partitions through calibration, thus obtaining location information such as "dent - left middle" and "bulge - top cover center" that can be used for subsequent retrieval.

[0074] Subsequently, based on defect type and location, a targeted search is performed in the locally stored target historical operation summary data to find stress event summaries that occurred within a preset time window and whose amplitude exceeds a preset physical threshold. The preset time window can be aligned with the transportation process, for example, from the estimated delivery time to the arrival time at the work site, or several hours / days before arrival at the work site; the preset physical threshold can be expressed in engineering-executable dimensions, such as an acceleration impact peak exceeding Xg, vibration intensity exceeding Y, or the number of impact events exceeding N. The stress event summary must at least include the event occurrence time and event amplitude; for example, the impact peak recorded by the accelerometer is summarized and written into the historical operation summary data.

[0075] When a stress event summary that meets the preset matching conditions is retrieved, the target battery to be treated is determined to have an internal damage risk. The preset matching conditions can include at least time matching and amplitude matching, and optional conditions can be added to improve accuracy, such as the defect location being consistent with the impact direction / stressed part, multiple impacts occurring within the same time period, or abnormal temperature rise occurring immediately after an impact.

[0076] When the matching conditions are met, statistical analysis results containing a high-risk safety level and anomaly interception instructions are directly generated, triggering the sorting system to transfer the battery to the anomaly review channel or isolation disposal area, preventing batteries with potential safety hazards from entering subsequent cascade utilization stages. To reduce the risk of missed detections, when no matching stress event summary is found, the main analysis process related to electrical performance data can continue to be executed, and the final diversion instruction is given from the comprehensive results.

[0077] To further improve the accuracy of the assessment, especially in identifying potential losses that may occur during transportation, this method can also incorporate the analysis of transportation environment data.

[0078] In another embodiment of this application, the method further includes: S6000-1: Obtain the transportation trajectory timeline of multiple batteries to be processed, which includes the location and time information of the multiple batteries to be processed; S6000-2: Before the target battery arrives at the work site, access third-party meteorological data sources or vehicle sensor records based on the transportation trajectory timeline to generate corresponding transportation environment summary data for the target battery. S6000-3: After analyzing the historical operation summary data and real-time detection data of the target to obtain the statistical analysis results of the target battery to be processed, it also includes: S6000-4: Based on the transportation environment summary data and the target historical operation summary data, determine the prediction reference range for consistency judgment, and perform consistency judgment between the real-time detection data and the prediction reference range to obtain the consistency judgment result; S6000-5: When the consistency determination result is within the preset consistency threshold range, the target battery to be processed is determined to be in a state that meets expectations during transportation, and the statistical analysis results are corrected based on the consistency determination result. S6000-6: When the consistency determination result exceeds the consistency threshold range, the statistical analysis result will be marked as transportation loss pending.

[0079] First, obtain the transportation trajectory timeline of multiple batteries to be processed. The trajectory data can be exported from the trajectory interface of the logistics platform or reported by the vehicle positioning terminal. Common fields include timestamp, latitude and longitude, vehicle speed, station events (loading / transfer / arrival), etc.; for example, latitude and longitude are recorded every 5 minutes to form a trajectory sequence sorted by time.

[0080] Before the target battery arrives at the work site, a pre-fetching mechanism is used to access third-party meteorological data sources or vehicle-mounted sensor records based on the transportation trajectory timeline. For example, environmental data such as temperature, humidity, and rainfall can be obtained by querying the meteorological service interface according to the latitude, longitude, and timestamp of the trajectory point; sensor data such as vibration, impact, and internal temperature can also be read from the vehicle data recording unit. Subsequently, the raw environmental data is summarized to generate transportation environment summary data, which includes, for example, the highest and lowest temperatures during transportation, the temperature variation range, the cumulative duration of temperature exceeding the preset temperature threshold, as well as the peak impact, number of impacts, or average vibration intensity. The summary results are then stored locally after being associated with the battery identification information for subsequent consistency determination.

[0081] After analyzing the target's historical operational summary data and real-time monitoring data and obtaining statistical analysis results, a consistency determination is introduced to identify potential losses that may occur during transportation. The consistency determination first determines a predictive reference range based on the transportation environment summary data and the target's historical operational summary data. This predictive reference range can be generated using rules, such as selecting a set of allowable offsets based on a pre-defined lookup table rule using "cumulative throughput range + maximum transportation temperature range + cumulative over-temperature duration range"; or mapping the maximum transportation temperature, over-temperature duration, and impact peak value to expected change ranges for indicators such as internal resistance and open-circuit voltage using a linear / regression model, thereby obtaining a reasonable range for the real-time monitoring data.

[0082] Subsequently, the real-time detection data collected on-site is compared with the predicted reference range to determine the consistency, and the consistency determination result is obtained. The consistency threshold range can be configured in an engineering manner, for example, setting the internal resistance to "fall within the predicted range or the Z-score does not exceed the threshold", and setting the voltage to "deviation not exceeding ±ΔV"; the threshold can also be configured according to battery model, expected arrival batch or season to adapt to the dispersion of batteries from different sources.

[0083] When the consistency determination result falls within the consistency threshold range, it can be determined that the transportation process status meets expectations, and the statistical analysis results can be corrected based on the consistency determination result, such as increasing the confidence level of the analysis results, reducing the weight of anomalies caused by transportation factors, or marking the transportation consistency as passed in the report.

[0084] When the consistency judgment result exceeds the consistency threshold range, the statistical analysis result can be marked as transportation loss pending, and a review prompt message can be output to guide further inspection or supplementary testing process to determine the cause of the deviation; for example, priority should be given to reviewing situations such as missing impact records, abnormal temperature inside the container, or long-term overheating caused by transit delays.

[0085] To achieve the above consistency determination, it is necessary to efficiently generate transportation environment summary data. The specific process for generating transportation environment summary data for the corresponding target battery to be processed is as follows: In another embodiment of this application, it is further proposed to generate transport environment summary data corresponding to the target battery to be processed, including: S6000-21: Divide the transportation process into multiple consecutive time periods based on the transportation trajectory timeline; S6000-22: For each time period, based on the location and time information corresponding to the transportation trajectory time axis, environmental data corresponding to that time period is obtained from third-party meteorological data sources or vehicle sensor records. S6000-23: Perform summary processing on environmental data to generate transportation environment summary data. The transportation environment summary data shall include at least one or more of the following: the highest temperature during transportation, the lowest temperature, the temperature change range, and the cumulative duration during which the temperature exceeds a preset temperature threshold. S6000-24: After associating the transportation environment summary data with the corresponding battery identification information, store it in local storage for consistency determination.

[0086] First, based on the transportation trajectory timeline, the transportation process is divided into multiple consecutive time periods. Time periods can be divided according to fixed time windows, such as intervals of 1 hour; or they can be divided according to trajectory events or area boundaries, such as when a vehicle enters / leaves a geofence. However, each segment is still defined by start and end timestamps to ensure the continuity of the time periods.

[0087] Next, for each time period, environmental data is obtained based on the latitude, longitude, and timestamp covered by that time period. Third-party meteorological data can be queried using "latitude, longitude, and timestamp" to find the nearest meteorological station or grid data; when the interface returns a coarse-grained time value, the nearest value or linear interpolation can be used to align to the time period. Vehicle-mounted sensor records can be filtered by timestamp to select data points falling within the interval, such as box temperature, vibration intensity, and impact peak value, and engineering rules can be applied to fill missing points or remove outliers.

[0088] After obtaining the raw environmental data, it is summarized to generate transportation environment summary data. The summarization process can be implemented using a "segmented statistics → full-process aggregation" approach: first, the highest temperature, lowest temperature, and duration exceeding the threshold are calculated for each time period; then, all time periods are aggregated to obtain the overall highest temperature (the maximum value of the highest temperatures across all segments), the overall lowest temperature (the minimum value of the lowest temperatures across all segments), the temperature variation range (the difference between the overall highest temperature and the overall lowest temperature), and the cumulative duration of temperatures exceeding a preset temperature threshold (the sum of the durations exceeding the threshold for each segment). For example, if the preset temperature threshold is 45℃, the cumulative number of minutes with temperatures above 45℃ throughout the transportation process can be counted to characterize the intensity of heat exposure.

[0089] Finally, the transportation environment summary data is associated with the battery identification information and stored locally. For multiple batteries in the same train or container, the transportation trajectory timeline and environment summary data are usually the same. Therefore, transportation environment summary data at the train / container level can be generated first, and then associated with each battery identification information, thereby reducing redundant calculations and storage. During subsequent consistency determination, the corresponding transportation environment summary data can be quickly located based on the battery identification information, providing data support for constructing the prediction reference range.

Claims

1. A method for automatic data collection and statistical analysis throughout the entire data lifecycle, characterized in that, The method includes: Obtain battery identification information and estimated arrival time of multiple batteries to be processed, including the target battery to be processed; Based on the estimated arrival time, before the battery to be processed arrives at the work site, a pre-fetching task is automatically triggered. Using the identification information of each battery as the retrieval key, the remote data source is accessed according to the preset supplier access table to obtain the historical operation summary data formed during the service life of the corresponding battery to be processed. The historical operation summary data is then associated with the corresponding battery identification information and stored in local storage. The historical operation summary data includes a set of agreed summary values ​​returned by the remote data source. In response to the detection that the target battery to be processed has arrived at the work site, the target battery to be processed is identified to obtain its battery identification information; Using the battery identification information of the target battery to be processed as the search key, the corresponding historical operation summary data is retrieved from the local storage as the target historical operation summary data; Real-time detection data of the target battery to be processed is automatically collected at the work site; The target historical operation summary data and the real-time detection data are analyzed to obtain the statistical analysis results of the target battery to be processed. The statistical analysis results include at least one or more of the following: residual value level, safety level, and shunt command.

2. The method for automatic data collection and statistical analysis throughout the entire lifecycle as described in claim 1, characterized in that, Automatically trigger a prefetching task, using the battery identification information as the retrieval key, and access remote data sources according to a preset supplier access table, including: For each battery identification information and its corresponding estimated arrival time, the difference between the estimated arrival time and the current system time is calculated to determine the remaining time window; When the remaining time window is greater than a preset time threshold, the battery identification information is added to the first preprocessing queue, and the remote data source is accessed according to the preset first resource configuration parameters; When the remaining time window is less than or equal to the preset time threshold, the battery identification information is added to the second preprocessing queue, and the remote data source is accessed according to the preset second resource configuration parameters; The first resource configuration parameter and the second resource configuration parameter include at least the maximum number of concurrent connections, the request timeout time, and the queue priority identifier, and the priority of the second resource configuration parameter is higher than the priority of the first resource configuration parameter.

3. The method for automatic data collection and statistical analysis throughout the entire lifecycle as described in claim 1, characterized in that, Analyzing the target's historical operational summary data and the real-time detection data yields statistical analysis results for the target battery to be processed, including: At least one electrical performance indicator to be tested is determined based on the real-time detection data; Based on the target's historical operation summary data, a historical reference index corresponding to the electrical performance index to be tested is determined. The historical reference index includes the summary characterization value of the electrical performance index to be tested during its service period or its reference range. The electrical performance index to be tested is compared with the historical reference index to obtain the historical comparison result; Obtain the testable electrical performance index samples of the remaining batteries that belong to the same expected arrival batch and have been tested as the target battery from multiple batteries to be processed, and calculate the dynamic statistical benchmark, which includes at least the sample mean and sample standard deviation. Based on the degree of deviation between the electrical performance index to be tested and the sample mean, and after normalization according to the sample standard deviation, the standardized deviation value of the target battery to be processed is calculated, and based on the electrical performance index to be tested and the sample mean, the absolute deviation of the target battery to be processed is calculated. Based on the index type of the electrical performance index to be tested and the batch information corresponding to the target battery to be processed, the standardized deviation judgment threshold and the absolute deviation judgment threshold are obtained from the preset judgment threshold table. Based on the index type of the electrical performance index to be tested, the historical comparison judgment threshold is obtained from the judgment threshold table; When the absolute value of the standardized deviation is less than the standardized deviation judgment threshold, or the absolute deviation amount is less than the absolute deviation judgment threshold, and the historical comparison result meets the historical comparison judgment threshold, a grade analysis report of the target battery to be processed is generated as the statistical analysis result. When the absolute value of the standardized deviation is greater than or equal to the standardized deviation judgment threshold and the absolute deviation is greater than or equal to the absolute deviation judgment threshold, or when the historical comparison result does not meet the historical comparison judgment threshold, an anomaly interception instruction is generated as the statistical analysis result.

4. The method for automatic data collection and statistical analysis throughout the entire lifecycle as described in claim 1, characterized in that, Using each battery identification information as a search key, a remote data source is accessed according to a preset supplier access table to obtain the historical operation summary data generated during the usage period of the corresponding battery to be processed. The historical operation summary data is then associated with the corresponding battery identification information and stored in local storage, including: The supplier access table is queried, and the corresponding remote data source access parameters are determined based on each battery identification information. The remote data source access parameters include at least the interface address, authentication method, and data format. A data request is sent to the remote data source according to the remote data source access parameters. The data request is used to instruct the remote data source to return a set of agreed-upon summary values. The set of agreed-upon summary values ​​includes at least the quantile description of the voltage rebound characteristic index, the historical maximum temperature rise, and the cumulative throughput. Based on the preset field mapping rules and value normalization rules in the supplier access table, the agreed summary value set is standardized to obtain standardized historical operation summary data. The standardized historical operation summary data is associated with the corresponding battery identification information and stored in the local storage. The standardized historical operation summary data is associated with the timestamp, version number and signature verification value to verify the timeliness and completeness of the standardized historical operation summary data.

5. The method for automatic data collection and statistical analysis throughout the entire lifecycle as described in claim 3, characterized in that, The method further includes: When retrieving the target historical runtime summary data from the local storage fails, or when the retrieved target historical runtime summary data is incomplete, a degradation statistical analysis process is initiated: Obtain real-time electrical performance parameters obtained by performing a short-time pulse load test on the target battery to be processed; The deviation is determined based on the real-time electrical performance parameters and the dynamic statistical benchmark. Query the preset threshold adjustment table to obtain the threshold judgment threshold corresponding to the expected arrival batch to which the target battery to be processed belongs; When the deviation is greater than or equal to the threshold judgment threshold, an abnormal interception command is generated as the statistical analysis result of the target battery to be processed; When the deviation is less than the threshold, a downgrade level analysis report is generated as the statistical analysis result of the target battery to be processed.

6. The method for automatic data collection and statistical analysis throughout the entire lifecycle as described in claim 4, characterized in that, The method further includes: When the supplier access table does not match the battery identification information, or the remote data source does not return the agreed summary value set within the preset waiting time, a general access and parsing process is executed: a general data request is sent to the remote data source to obtain response data; according to the preset general field position rules, the response data returned by the remote data source is processed to extract fields and obtain field values; according to the preset minimum unit conversion rules, the field values ​​are converted and formatted to generate at least one historical operation summary data item to constitute the historical operation summary data; and the historical operation summary data is associated with the corresponding battery identification information and stored in the local storage. When the general access and parsing process fails to obtain the historical operation summary data that meets the preset integrity requirements, a field compensation mark is generated for the corresponding battery identification information. After the target battery to be processed arrives at the work site, the detection equipment is controlled to perform a short-time pulse load test on the target battery to be processed, so as to supplement the generation of field compensation historical operation summary data items. Based on the field compensation historical operation summary data items, the historical operation summary data associated with the battery identification information and bearing the field compensation mark is compensated and updated to obtain the compensated historical operation summary data. After establishing an association between the compensated historical operation summary data and the battery identification information, it is updated and stored in the local storage.

7. The method for automatic data collection and statistical analysis throughout the entire lifecycle as described in claim 1, characterized in that, Analyzing the target's historical operational summary data and the real-time detection data yields statistical analysis results for the target battery to be processed, including: Obtain packing rules for multiple batteries to be processed, wherein the packing rules are used to indicate the loading topology of the batteries to be processed within the transport container; According to the packing rules and the battery identification information of the target battery to be processed, the target battery to be processed is mapped to the hot zone inside the transport box. The hot zone includes at least the edge zone, the transition zone and the core zone. The expected temperature deviation level of the target battery to be processed is determined based on the thermal zone using a preset temperature response coefficient table. Based on the temperature response coefficient table and the expected temperature offset level, the real-time detection data is normalized and corrected to obtain temperature-corrected real-time detection data. The statistical analysis results of the target battery to be processed are obtained by analyzing the real-time detection data after temperature correction and the target's historical operation summary data.

8. The method for automatic data collection and statistical analysis throughout the entire lifecycle as described in claim 1, characterized in that, The real-time detection data includes visual appearance data. The analysis of the target's historical operational summary data and the real-time detection data to obtain the statistical analysis results of the target battery to be processed includes: When the visual appearance data indicates that the target battery to be processed has a physical defect, the type and location of the physical defect are determined; A targeted search is performed in the target historical operation summary data to find stress event summaries that occurred within a preset time window and whose amplitude exceeded a preset physical threshold. The stress event summaries include at least the stress event occurrence time and stress event amplitude. When a stress event summary that matches the physical defect with a preset matching condition is retrieved, it is determined that the target battery to be processed has an internal damage risk, and a statistical analysis result containing a high-risk safety level and an abnormal interception command is generated.

9. The method for automatic data collection and statistical analysis throughout the entire lifecycle according to claim 1, characterized in that, The method further includes: Obtain the transportation trajectory timeline of multiple batteries to be processed, the transportation trajectory timeline including the location and time information of multiple batteries to be processed; Before the target battery arrives at the work site, the transportation environment summary data of the target battery is generated by accessing third-party meteorological data sources or vehicle sensor records based on the transportation trajectory timeline. After analyzing the target's historical operational summary data and the real-time detection data to obtain the statistical analysis results of the target battery to be processed, the method further includes: Based on the transportation environment summary data and the target historical operation summary data, a predictive reference range for consistency determination is determined, and the real-time detection data is compared with the predictive reference range to obtain a consistency determination result. When the consistency determination result is within the preset consistency threshold range, the target battery to be processed is determined to be in a state that meets expectations during transportation, and the statistical analysis results are corrected based on the consistency determination result. When the consistency determination result exceeds the consistency threshold range, the statistical analysis result will be marked as transportation loss pending.

10. The method for automatic data collection and statistical analysis throughout the entire lifecycle as described in claim 9, characterized in that, Generate a summary of the transportation environment data for the target battery to be processed, including: The transportation process is divided into multiple consecutive time periods based on the transportation trajectory timeline. For each time period, based on the location and time information corresponding to the transportation trajectory time axis, environmental data corresponding to that time period is obtained from the third-party meteorological data source or the vehicle-mounted sensor records. The environmental data is summarized to generate the transportation environment summary data, which includes at least one or more of the following: the highest temperature during transportation, the lowest temperature, the temperature change range, and the cumulative duration during which the temperature exceeds a preset temperature threshold. The transportation environment summary data is associated with the corresponding battery identification information and stored in the local storage for use in the consistency determination.

Images